Find highest quality duplicated instead of removing duplicates

I would like to do something similar to whats described in the docs, but instead of deleting duplicate files, I would like to search for duplicates(from a set a query images), find the duplicate with the highest quality, and  copy that duplicate to a new folder